Abstract

Brian C. Stuart1

Brian C. Stuart1

One of the responsibilities of the International Monetary Fund in promoting international monetary cooperation is the provision of financial support to the adjustment programs of countries suffering external payments difficulties. The IMF’s loans are not for specific development projects but represent general balance of payments financing. Resources are provided in support of macroeconomic adjustment and structural reform programs under stand-by arrangements generally covering one or two years, and the IMF supports comprehensive macroeconomic and structural adjustment programs under three-year extended arrangements (which under certain circumstances can be lengthened to a fourth year). The rate of interest on these resources is market related; the repayment period is generally three to five years for resources borrowed under stand-by arrangements and up to ten years for extended arrangements.

In addition, under the structural adjustment facility and the enhanced structural adjustment facility (SAF/ESAF), the IMF administers concessional loans to low-income developing countries in support of three-year structural adjustment programs. (Under the ESAF, arrangements can be extended to a fourth year.) The rate of charge on these loans is 0.5 percent a year and the repayment period is up to ten years.

The IMF also makes resources available to members under the compensatory and contingency financing facility (CCFF) to compensate for temporary shortfalls in exports of goods and services, to help members face unexpected changes in the external economic environment, and, on a temporary basis in the wake of the recent Middle East crisis, to compensate for temporary increases in oil import costs. Finally, since 1989 the IMF has provided financial support to countries agreeing to debt-reduction operations with commercial bank creditors, in association with stand-by and extended arrangements.

As more countries are adopting macroeconomic adjustment and structural reform programs, the IMF is involved in lending to a record number of countries. At present, the IMF has arrangements with almost 50 member countries (about one-third of the total membership), representing commitments of about $22 billion. In addition, discussions on adjustment programs that could be supported by the IMF are being held with 15-20 other countries. Total outstanding IMF credit now amounts to about $34 billion.

In addition to supporting adjustment through the provision of financing, the IMF aims to improve the working of the international monetary system through its surveillance function. This involves regular visits by IMF staff to all member countries for discussions of economic policy with government officials—usually on an annual basis. This is followed by a staff report on economic conditions in the country and a full discussion of the member country’s economic policy in the IMF’s Executive Board.

The IMF also examines how economic policies in individual countries interact to affect the smooth functioning of the world economy. This analysis forms part of the World Economic Outlook exercise which is conducted twice a year. In addition, for meetings of the finance ministers and other officials of the seven largest industrial countries (the Group of Seven), the IMF analyzes the progress being made in reducing economic imbalances among the industrial countries and suggests policy adaptations that may be required.

Finally, the IMF helps member countries improve policy design through the provision of technical assistance. Earlier in the IMF’s history, such advice was limited to banking and monetary matters, but now the IMF provides assistance in many fields—fiscal, monetary, balance of payments, banking, exchange and trade systems, government finance, and statistics. Assistance is provided through staff missions, longer-term field assignments by staff members or outside experts, and studies prepared at the IMF’s headquarters.

Overview of the Adjustment Process

Imbalances and Structural Problems

Macroeconomic imbalances reflect an excess of domestic demand over supply and, often, serious distortions in the economy. Many developing countries are still dealing with the legacy of unsustainable levels of external borrowing in the late 1970s and the world recession in the early 1980s. For most countries, the external environment in the past several years has been mixed. While the world economy and trade expanded at impressive rates throughout the latter part of the 1980s, many countries experienced a downward trend in the relative price of their exports, and world interest rates remained at high levels in real terms. In addition to the large number of countries that adopted adjustment measures in the 1980s, in the past two years the countries of Eastern Europe have undertaken fundamental transformations of their economic systems.

Each country in need of adjustment has its own set of difficulties, built up over time in response to changing economic circumstances and particular pressures from various groups in the society. High rates of credit and money expansion and shortages of national savings often reflect large public sector deficits. Various distortions can reduce output and limit economic growth: quantitative barriers to imports and high import duties; inefficient public enterprises and distortionary tax systems that feature a narrow tax base with high rates; financial regulations that result in negative real interest rates, reduce saving, misallocate resources, and lead to a substitution of local currency by foreign currencies; consumer subsidies, price controls, and other impediments to the functioning of goods markets; and wage policies (including indexation and minimum-wage laws) and labor market regulations and conditions (limits on hirings and firings as well as weaknesses in the educational system and housing infrastructure) that limit the mobility of labor.2

Adjustment

In considering the adjustment process, it is important to recognize that a country cannot live beyond its means indefinitely. When imbalances exist, adjustment of one kind or another will take place. This adjustment can be orderly or disorderly, and an important role of the IMF is to help members ensure that the adjustment is as orderly as possible.3

Disorderly adjustment can manifest itself in shortages of goods and the emergence of black markets, in high rates of inflation (an inefficient and regressive form of taxation), and in the buildup of external payments arrears (a limited and very costly form of foreign finance). Orderly adjustment, on the other hand, is associated with declining inflation, the correction of economic distortions, and the prospect—at least in the medium term—of sustained economic growth. In general terms, the IMF assists in the design of orderly adjustment programs that help the borrowing country attain a viable balance of payments, sustainable economic growth, and reasonable price stability. At the same time, financing is provided to ease the adjustment.

In the past, IMF-supported programs tended to focus on the demand side of the economy attempting to bring aggregate demand in line with aggregate supply. The IMF worked on reducing the fiscal deficit, limiting overall credit expansion, and containing the balance of payments deficit and external borrowing. Moreover, the IMF’s formal conditionality on quantitative targets for key macroeconomic variables was also motivated by demand considerations. However, structural measures to increase supply were not ignored. All programs required the avoidance of trade and payments restrictions and other measures “destructive of national and international prosperity” (as required by Article I of the IMF’s charter). In this context, many programs featured adjustment of the exchange rate to ensure the proper alignment of domestic and foreign prices.

However, since the mid-1970s there has been an increasing recognition that the problems of some members are more intractable than originally thought, given a world economic environment that has often created difficulties for developing countries. This change in the IMF’s focus was also induced by the large number of developing countries joining the IMF with problems that required supply-side attention. The approach to economic adjustment has evolved accordingly and is reflected in a lengthening of the possible duration of adjustment programs supported by the IMF from one year to three years and most recently to a possible four years.

More important, from the point of view of this paper, the scope of issues stressed in adjustment programs has broadened considerably in recent years. Greater attention is being paid to the efficiency with which existing resources are used and to increasing the capability of economies to adapt to changing circumstances. These efforts involve the economic pricing of goods and of factors of production as well as improved management and possible privatization of public enterprises.

Attention is also being paid to the adverse effects of high tax rates on the public’s willingness to work, save, and invest and to the need to increase saving in the economy through strengthening public sector saving, establishing appropriate interest rate levels, and strengthening financial institutions. The liberalization of restrictions on foreign direct investment and other measures to deregulate the economy, as well as the establishment of clear rules of the game for foreign and domestic investors, has led to the increased availability of foreign savings and technology. Finally, efforts are being made to ensure more productive investment through better pricing and interest rate policies and careful setting of public sector investment priorities.

This greater focus on supply-side policies in the work of the IMF occurred at the same time that the World Bank was moving into the area of policy-based lending. These developments have been a factor leading to the intensified collaboration between the IMF and the World Bank.

Of course, the greater emphasis on addressing supply-side problems does not imply that the adjustment process is painless. First, in a situation where a country has been living on excessive levels of foreign borrowing or inflationary financing, spending often needs to be reduced, even though the society will tend to resist such attempts to bring overall spending into line with available resources. Second, efforts to improve the efficiency of the economy—to benefit society as a whole by raising overall levels of economic welfare—will be resisted by those groups that have enjoyed the extra income or rents associated with existing economic distortions.

It must be acknowledged that an important and frequent effect of adjustment programs is a change in the distribution of income and a lower level of spending by certain groups in society. It is also recognized that the only way to alleviate poverty in a lasting way is through the sustained growth of output in a stable macroeconomic environment.

However, it is important also to protect the poorest groups in society from possible short-run effects of adjustment. In its policy dialogue with member countries, the IMF is paying increasing attention to improving its knowledge of poverty in individual countries, to identifying the possible effects of adjustment programs on the poor, and to helping with the design of targeted programs to help shield the poor from adverse transitional effects of adjustment measures. In addressing poverty concerns, the IMF cooperates closely with the World Bank, other specialized United Nations agencies, and social ministries and nongovernmental organizations in individual countries.

Analyzing the Effects of Programs

Before turning to the experience of countries with macroeconomic and structural reform programs, it would be useful first to address important methodological points related to the assessment of adjustment programs. This is a difficult area and the issues remain controversial.

A number of studies have attempted to evaluate the effects of IMF supported programs on major macroeconomic variables, such as growth, inflation, and the balance of payments. More recently, studies have examined the effects of reform programs on income distribution and poverty.

In the studies, various approaches have been followed, each with its own strengths and weaknesses. Two shortcomings of most approaches are the lack of a method for “predicting” the policies that would have been pursued in the absence of the IMF-supported program and the absence of a model of economic adjustment that would permit the analysis of the effects of alternative policies.4 This problem is referred to as the absence of a “counterfactual” that would permit a detailed assessment of the adopted policy package in light of the alternative measures that could have been implemented.

Three main approaches to the evaluation of programs can be identified—the “before-after” approach, the “with-program and without–program” approach, and the “target versus actual” approach. The before-after approach compares developments in the macroeconomic variables in the period prior to the program with developments during and after the program. This is a fairly easy test to design. Its major shortcoming is that it does not take account of changes in nonprogram factors—such as terms of trade, world demand, and weather conditions. The approach thus attributes any change in economic performance between the pre- and post-program periods to the program itself. Also, it may be the case that the pre-program situation would have been unsustainable under any policy approach—for example, when high levels of spending are supported by excessive levels of external borrowing. In this context, the before-after approach incorrectly blames a deterioration in the economic situation on the adoption of adjustment measures rather than on the initial conditions facing the economy.

A comparison of a group of countries undertaking IMF-supported programs with a group of countries without such programs—the with–without approach—may get around some of the problems arising from changes in nonprogram factors (provided that these nonprogram factors affect both groups of countries in similar ways). However, the simple with-without approach does not take into account the different starting positions of the two groups of countries. It generally could be expected that countries entering into IMF-supported programs start from a weaker position than other countries, and a comparison of the two groups may incorrectly attribute poorer relative performance to the IMF-supported program.

As with the before-after approach, the with-without studies generally divide countries into two groups based on whether they have a program in place or not. As such, the comparisons do not take account of the extent to which programs have been implemented as planned and thus may not represent an evaluation of the programs as originally designed.

The third approach to program assessment is to compare the outturns for various economic variables with the objectives set out in the adjustment programs. This is an important comparison for an institution like the IMF to make when evaluating the effectiveness of the programs it has supported. However, this assessment of adjustment programs will be affected by the ambitiousness of the targets themselves. Also, as with the before-after approach, nonprogram factors can also affect outcomes independent of the quality of the program itself.

The methodological issues concerning the macroeconomic analysis of programs apply equally to an assessment of the programs’ effects on income distribution. The problems are compounded in a major way by a serious lack of information in most countries on income distribution and on the sources of income and the spending patterns of the poor.

Country Experiences

A number of studies on the effects of IMF-supported programs were surveyed in a recent paper by Mohsin Khan and can be summarized as follows.5 Almost all of the studies conclude that the programs were successful in containing inflation, while a somewhat smaller number of them conclude that the programs were helpful in reducing external current account or overall balance of payments deficits. Most of the studies conclude that, at least in the short run, the effects of the programs on economic growth were negative.

In his paper, Khan also examined the adjustment experiences of about 75 developing countries over the period 1973-88, correcting for some of the shortcomings in the standard approaches to program evaluation that were described above by taking account of certain nonprogram factors and differences in the starting positions of the adjusting countries. However, it was not possible to differentiate between countries based on whether programs were fully implemented or not. The results of Khan’s analysis were broadly consistent with earlier studies. In the first year of the programs, adjustment was seen as improving the external accounts and containing inflation, while the effects on growth were negative. When the analysis was extended to cover the first and second year of the program, the effect of adjustment on the external accounts and inflation was strengthened, while the negative effect on growth was lessened.

As part of an ongoing review, the IMF has recently examined the effects of programs in low-income countries supported by the structural adjustment and enhanced structural adjustment facilities (SAF/ ESAF), the macroeconomic effects of programs in middle-income countries supported by IMF stand-by and extended arrangements, and the implications of IMF-supported adjustment programs for poverty.

Recent SAF/ESAF Programs

The review of SAF/ESAF programs covered those in place between January 1990 and April 1991. As regards growth performance under these programs, about half of the programs achieved or exceeded their growth objectives, and output rose by almost 4 percent a year (Table 1). By way of comparison, output growth in these countries averaged about 2.6 percent a year in the three years preceding the program, and output growth for all of the least developed countries averaged 2.5 to 3 percent a year in 1990 and 1991.

Table 1.

Summary of Program Developments

article image
Source: IMF staff estimates.

The programs were somewhat less successful in meeting inflation targets. Only about a third of the programs met their inflation objectives and the median rate of inflation increased slightly from preprogram levels. In particular, it proved difficult to reduce inflation rates to below 10 percent a year.

In contrast to inflation performance, well over half of the programs met the external current account and overall balance of payments objectives. On average, the programs targeted a small increase in the current account deficit to allow for increased imports in light of increased foreign financing. In general, the weakening in the current account balance was contained to the targeted amount.

Stand-by and Extended Arrangements, 1985-88

The review of programs supported by stand-by and extended arrangements covered the period 1985-88. The objective for economic growth was met in about half of the programs, as were the objectives for inflation, the external current account, and the overall balance of payments. The growth rate under these programs averaged about 2.5 percent a year—roughly in line with targeted growth and well above the average growth rate in the pre-program year of just over 1 percent.

Inflation was brought down on average, but, as with SAF/ESAF programs, the rate of price increase remained above targeted levels. Particular problems were encountered in meeting the inflation objectives in cases where the rate of price increase was targeted to remain relatively high in the program year. On average, the targeted improvement in the external current account was achieved, but the objective for the improvement in the overall balance of payments was not.

To review programs supported by stand-by and extended arrangements, an attempt was made in Table 2 to evaluate the effects of program implementation on economic performance. The results suggest a fairly strong association between the implementation of fiscal and credit policies and program results. Objectives with respect to growth, inflation, and the balance of payments were met in 60-75 percent of the cases when both credit and fiscal policies were implemented as planned. These objectives were met in fewer than half of the cases when neither credit nor fiscal targets were met. For programs meeting both fiscal and credit targets, the rate of economic growth averaged 4.4 percent a year, well above the targeted rate (Table 3).

Table 2.

Program Implementation and Results

(Number of programs)

article image
Source: IMF staff estimates.
Table 3.

Program Implementation and Macroeconomic Performance

article image
Source: IMF staff estimates.

If the two programs that targeted an inflation rate of 40 percent or more are excluded, the averages were 13 percent (pre-program) and 11 percent (target and actual).

The evidence from the programs suggests a strong association between developments in the terms of trade and a country’s ability to implement adjustment policies as planned. In 11 of the 22 programs where terms of trade movements met or exceeded projections, both credit and fiscal targets were met. In contrast, only 3 of the 16 programs that experienced weaker than expected terms of trade met both credit and fiscal targets. There was also a close relationship between terms of trade developments and performance with respect to the external current account. In almost all of the cases where the terms of trade met or exceeded expectations, the current account target was met.

The association between the implementation of fiscal and credit policies and the achievement of inflation and external objectives suggests that the approach to financial programming has been broadly appropriate. The possible linkages between fiscal and credit policies and growth performance are more complex. The correlation between program implementation and higher than projected growth could suggest that adjustment supports economic growth, even in the short run. The results also point to the importance of a balanced approach to adjustment. Credit policy was relatively effective in containing inflation even in those cases where fiscal policy was not implemented as planned, but this mixed approach was associated with weaker than projected economic growth.

Care is needed in drawing conclusions about cause and effect relationships based on the program-versus-actual approach. However, the growth performance of countries under recent IMF-supported programs is encouraging. It suggests that the approach to adjustment adopted in recent years, involving a comprehensive set of macroeconomic and structural measures, is supportive of growth. At the same time, the recent drop in output in the Eastern European countries, as these economies adopt comprehensive reforms, suggests that the transitional costs of a fundamental transformation of economies—short-run declines in output and employment—may be substantial.

Effects of Adjustment Programs on the Poor

A recent study by the IMF of the effects of adjustment programs on the poor, based on case studies of selected countries, concluded that many adjustment measures directly benefit the poor—currency depreciation raised the incomes of the rural poor (Ghana, Kenya, and the Philippines); financial system reform increased the availability of credit to the poor (Ghana, Kenya, and Chile); and the removal of price controls reduced rents based on political privilege (Ghana and the Philippines).6 The paper also concluded that the poor are least able to protect themselves from the effects of inflation, shortages, and other consequences of disorderly adjustment and that the poor may consequently be better off in a period of orderly adjustment.

At the same time, the study acknowledged that some adjustment measures can hurt the poor. Examples were the adverse effects of currency devaluation on the urban poor or in a nontradable goods sector that was labor intensive (Chile, the Dominican Republic, and the Philippines), cuts in health and education expenditure accruing to the poor (Sri Lanka and the Philippines), and restrictive monetary policies affecting employment (the Philippines).

As noted above, the IMF is paying greater attention to the social aspects of adjustment and an increasing number of programs include specific social measures, particularly the programs of the poorest member countries supported by the IMF’s structural adjustment and enhanced structural adjustment facilities. These measures fall into four main categories—consumer subsidies of certain goods; income transfers from the budget; wage policies, job programs, and other measures to directly raise the income of the poor; and protection of education and health expenditures.

Among the various adjustment programs, consumer subsidies have involved direct subsidies for certain goods consumed by the poor or indirect subsidies through an exemption of the sales tax on basic foodstuffs as well as the waiving of user charges in priority sectors, such as basic health and education. These efforts have been better targeted in recent programs.

Income transfers from the budget have varied and included direct financial support to some affected groups and a general reallocation of central government expenditure. Direct financial support has been provided to civil servants laid off in the wake of administrative reforms or to low-paid civil service employees after the removal of subsidies on the sale of basic foodstuffs. The reduction in central government transfers and subsidies has been accompanied by a change in its composition—increased direct transfers and special programs (food stamps and midday meal programs) to the poorest households.

A third group of actions has consisted of efforts to raise directly the income earned by the poor. In a number of programs, this action was undertaken through an increase in the minimum wage or the creation of employment opportunities, for example, through road repairs and irrigation projects.

Finally, most programs have recognized the need for increases in social expenditure for basic education and health and the need for poverty alleviation. In general, increases in this area were aimed at meeting basic needs as well as redressing inequities caused by expenditure cuts in the adjustment program. In most cases, this additional expenditure has been funded by a cut in other categories of expenditure, including budgetary subsidies and restraint in the growth of wages and salaries. Basic education and health have been identified as priority sectors in a number of programs.

Programming Issues

In designing adjustment programs, a number of important programming issues arise, including the linkages between macroeconomic stabilization and structural reforms, the appropriate pace and comprehensiveness of reforms, and the sequencing of reforms. The adjustment strategy that is chosen will depend on the particular circumstances of each country, but some general lessons are emerging to guide the choices facing country authorities. In the end, the success of the effort will depend crucially on the commitment of the government to the program and the perseverance shown in its implementation.

Macroeconomic Stability and Structural Reform

Macroeconomic stability can be seen as a prerequisite for sustained growth of output. In other words, there appears to be no trade-off, especially in the medium term, between output growth and inflation. When developing countries are divided into two groups depending on whether they have an inflation rate above or below the median, the countries with low inflation had an average national saving rate almost 10 percentage points of GDP higher than the high-inflation countries over the 1982-88 period; their growth rate averaged about 3.5 percentage points higher.7

It is also generally accepted that structural reform will be largely ineffective in a situation of financial instability. High rates of inflation mean that prices and other market signals will not be particularly useful to economic agents in deciding how to allocate resources. This suggests that macroeconomic adjustment should precede structural reforms or that steps to restore macroeconomic balance should be taken simultaneously with structural measures.

The lesson from partial market-oriented reforms in centrally planned economies in the 1980s—particularly China and Yugoslavia—is that once direct controls are relaxed a good deal of progress in the structural area may be needed to maintain macroeconomic stability. As more autonomy is given to enterprise managers and other economic agents, control over the demand and supply of credit and over wage policy may decrease unless deeper reforms of enterprises and the financial system are undertaken, leading to a hardening of the budget constraint facing producers. The recent experience of the countries in Eastern Europe confirms this linkage between the need for progress in the structural area and the establishment of macroeconomic stability.

Comprehensiveness and Sequencing of Reforms

A second issue to be addressed in the design of adjustment programs is the comprehensiveness and sequencing of reforms. The comprehensiveness of needed reforms depends on the extent of the distortions in the economy. Given the linkages between various reforms, a comprehensive approach is needed in an economy with pervasive distortions.

For example, reform of the exchange and trade system would have little effect if distortions in the domestic pricing system are not also removed. At the same time, price reform would have little effect on the allocation of resources unless enterprise reform is undertaken to give firms an interest in increasing profits (and make them responsible for their losses). Credit would not be allocated efficiently unless the financial system is reformed to strengthen the supervision of banks and give greater play to market forces. Tax reform would not necessarily improve efficiency if personal and business income is based on a distorted wage and price structure.

The reallocation of labor among different industries and geographical areas (and the related transitional costs of higher unemployment) can be smoothed by the establishment of social safety nets. In societies where family allowances, health benefits, and housing are provided by employers, it will be necessary to make the provision of these benefits and services independent of the place of employment in order to increase labor mobility.

Indeed, the theory of the second best suggests that the removal of a single distortion, while other distortions remain, may actually worsen economic welfare rather than improve it. For example, the decentralization of economic decision making in a situation of continued serious price distortions may result in a less efficient allocation of resources than the continued centralized control of resources. This argument strengthens the case for a comprehensive approach to reforms.

However, many countries lack the administrative capacity for a comprehensive approach, or governments may not have sufficient political support to undertake reforms simultaneously across a broad front. In these circumstances, some sequencing of reforms may be necessary. The ideal sequence depends on the particular circumstances of each case, but as a general rule it is considered appropriate to reform first those markets that are slowest to adjust or most costly to change. More specifically, labor and goods markets generally should be freed before financial markets. Thus, it is widely accepted that one of the problems encountered in the liberalization programs of the southern-cone countries (Argentina, Chile, and Uruguay) in the late 1970s and early 1980s was that financial markets and the capital account of the balance of payments were liberalized ahead of trade reform.

Following on this argument, financial system reform affecting the allocation of investment resources would not be efficient if it were attempted before a reform of domestic prices took place that ensured resources are moved into appropriate activities. Similarly, it would not be appropriate first to adjust the structure of domestic prices in order to clear domestic markets and then, only after the costly reallocation of resources had occurred, to open up the trade system for a further round of changes in production and consumption patterns based on world prices. Thus, in the recent reform programs in centrally planned economies, particular emphasis has been placed on the early reform of prices and the trade and exchange rate systems.

Speed of Adjustment and Reform

The debate over the speed of adjustment—a gradual approach versus shock therapy—is a long-standing one, to which there is no easy answer. However, the experience with shock approaches to the elimination of inflation in Argentina, Bolivia, Brazil, and Israel in the mid- 1980s, after the repeated failure of gradualist approaches, showed that inflation could be brought down quickly even in countries with a history of high inflation.

In these cases, output also recovered fairly quickly, after an initial period of weakness. This response of output has been attributed to either the supply-side effects associated with more efficient resource use in a low-inflation environment or the redistribution of income (associated with the elimination of the inflation tax) toward lowerincome families, who have higher propensities to spend. The shortlived success of the anti-inflation programs of Argentina and Brazil in the mid-1980s also points to the crucial importance of fiscal and monetary restraint if inflation is to be kept low.

Early experience with the bold adjustment programs that have been undertaken in Eastern Europe—where output has dropped more quickly than originally expected—suggests that the supply-side effects of more efficient resource use may not be sufficient, in the short run, to overcome the transitional drop in output associated with a major transformation of the economy. To some extent, the drop in output can be attributed to the collapse in trade with the former Soviet Union and other members of the recently dissolved Council of Mutual Economic Assistance and to delays in the implementation of reform measures. It may also be that the official data on output do not accurately reflect a pickup in private activity associated with the reforms. Nevertheless, it would seem that the degree of inefficiency in these economies and the transitional costs of economic adjustment were underestimated.

Quality Versus Quantity in Fiscal Adjustment

As adjustment programs give greater emphasis to structural aspects, particularly in the fiscal area, the possible trade-off between the quality and the quantity of fiscal adjustment becomes an issue.8 The approach followed by the IMF when discussing adjustment programs with member countries is to estimate the reduction in the fiscal deficit needed to achieve the macroeconomic objectives of the program.

Meeting these fiscal targets may preclude the reduction of inefficient taxes or require cuts in public investment or other productive expenditure.

Nevertheless, recent programs supported by the IMF have increasingly recognized that the specific measures through which fiscal adjustment is achieved have important effects on the medium-term success of the program. A question from the point of view of the IMF is the extent to which the emphasis on structural reform should be reflected in the conditionality attached to its lending operations—in other words, the extent to which continued access to IMF resources should be conditional on the quality of adjustment as well as on the meeting of macroeconomic targets.

Determinants of Program Implementation

The success of adjustment programs depends on careful program design, which, in turn, requires an accurate assessment of the problems facing the economy, a detailed specification of measures to be adopted, careful program monitoring, and protections for the program against the effects of adverse exogenous shocks. Such protections include the provision of additional financing and the adaptation of the program to the changed external economic environment. In the end, however, the degree of implementation will depend on the strength of the authorities’ commitment to the program.

This commitment can be strengthened by ensuring that the authorities are involved at the very early stages of problem diagnosis and program design and that all agencies of the government responsible for the implementation of the program are committed to the program. In the IMF and the World Bank it is said that the member country must “own” the program if it is to be successful. Very often, the success of a program is limited because support within the government is too narrowly based. Broad public support for a program can be fostered by clearly explaining the benefits to be gained from early and orderly adjustment and by being realistic about the timing of the expected benefits.

Concluding Remarks

The nature of programs supported by the IMF have changed through time. While the supply-side, or structural, aspects of adjustment were by no means ignored in the past, they have been given substantially more emphasis in recent programs.

Similarly, more focus has been given to the effects of adjustment on the distribution of income and on the poor. The IMF continues to deepen its understanding of these issues, while helping member countries design measures to protect the poor from the transitional costs of adjustment.

While care must be taken in drawing strong conclusions from the experiences of countries adopting adjustment measures, the recent assessment by the IMF of economic developments under programs it supports show that adjustment can be associated with growth even in the short run. However, the experience of the centrally planned economies in Eastern Europe suggests that a decline in output during major economic transformations may be inevitable. Performance under recent stand-by and extended arrangements also suggests that the financial programming approach to reducing inflation and containing external imbalances, which has been adopted in recent programs, is broadly appropriate.

The relationship between macroeconomic adjustment and structural reforms is being increasingly understood. Financial stability is seen as a prerequisite for the effectiveness of structural reforms; at the same time, in countries characterized by widespread market distortions, major progress in structural reform can be seen as essential to the establishment or maintenance of macroeconomic stability.

While the correct sequencing of reforms must be determined on the basis of a careful analysis of each case, general lessons about the appropriate sequencing of reforms are also becoming clear—the liberalization of goods and factor markets should precede the freeing of financial markets. In the final analysis, however, the success of adjustment programs will depend on the design of adjustment programs by the authorities, on a detailed understanding of the characteristics of their economies, and on their perseverance in implementing the programs.

Comments

Akbar Noman1

Brian Stuart does a commendable job of succinctly covering a wide gamut of issues in his ambitious paper, although he does not simply summarize issues and evidence. Along the way, Stuart takes positions in complex, ongoing debates, making for a stimulating essay but also a somewhat tendentious one, if for no other reason than that he seems cramped for space. Because of this scope, his is not an easy paper on which to comment. Thus, the parts that are persuasive or unexceptionable, as well as some of the caveats in the paper, I shall ignore in order to sharpen my focus on some of its bolder and more contentious aspects.

The author’s paper contains an excellent summary of the methodological problems involved in evaluating the effects of programs supported by the International Monetary Fund. These problems stem essentially from the absence of a counterfactual in making such assessments—that is, the inability to measure what would have happened in the absence of a given program. By now there is a fair body of studies that attempt to substitute for the counterfactual, and this evidence shows that programs have favorable effects, though the findings are neither overwhelming nor entirely conclusive. The question that arises, then, is when and to what extent do programs not meet their objectives and why do they fail? One of the following five reasons might provide an answer:

  • (i) problems in measuring the effects of programs (for example, poor national accounts), which create the illusion of failure;

  • (ii) exogenous shocks of external or domestic origin (terms of trade, political crises, or weather);

  • (iii) unrealistic expectations about the program’s likely results (forecasting the effects of a certain set of policies on such variables as economic growth, inflation, and the balance of payments is an imperfect art, and the forecasts may reflect wishful thinking);2

  • (iv) inadequate implementation; or

  • (v) faulty design.

The last two are closely related and may be difficult to distinguish: weak implementation may reflect design inadequacies. Thus, if programs do not adequately take into account the constraints on implementation, political feasibility, or “ownership” by governments, they are likely to suffer in implementation. Programs may not only neglect but even exacerbate these constraints by being overambitious, internally inconsistent, or overloaded with conditionality. There are difficult judgments to be made about what constitutes an appropriate, feasible, and adequate policy. Stuart’s paper is not unmindful of these considerations but in my view pays too little attention to them.

There is some tension in the paper between emphasizing the specific circumstances of a country and seeking to draw general lessons, with Stuart leaning toward the latter. In drawing general lessons, it would have been helpful to develop a typology of country circumstances or initial country conditions. As Stuart recognizes, an important question is how general these lessons truly are. Indeed much of the current debate on macroeconomic and structural adjustment turns on this question.

On the sequencing of macroeconomic stabilization and structural reforms, Stuart argues that there is no conflict between stabilization and growth, at least in the medium term. He also suggests that stabilization should take precedence over structural reforms except when major systemic reforms are being undertaken, such as in Eastern Europe. Yet there is also an argument for reverse sequencing. Stabilization and growth are both complements and substitutes: macroeconomic stability can facilitate sustained growth, but too strong a dose of stabilization can hinder growth. The issue is one of striking the right balance between supply-side and demand-side measures, a balance that depends on the initial country conditions. The question of the sequencing between stabilization and structural reforms arises only in acute cases of macroeconomic imbalance or major systemic reforms. In the case of the former, there is a strong presumption that stabilization should precede structural reforms. In the latter situation, the reverse holds true.

On the scope of programs, Stuart argues for comprehensiveness, invoking the theory of the second best, which says that the removal of a single distortion while other distortions remain may actually worsen economic welfare not improve it. However, the theory of the second best also says that removing, say, 20 out of 30 distortions is not necessarily better than removing 10 of them, nor does it state that a faster reduction in any given set of distortions is always better than a slower one.

Stuart acknowledges the constraints of administrative capacity and political support facing comprehensive reform. Stuart also points out that it is a good idea to liberalize slower-adjusting markets like the goods and labor markets before liberalizing the quick-adjusting ones like the financial market. But other good reasons also favor a less comprehensive approach in particular areas. For example, it has been argued that export promotion should precede across-the-board import liberalization, as distinct from selective liberalization, in order to ease the balance of payments constraint that such liberalization can quickly run into. Another important argument against more rather than less comprehensive reform is the capacity of economic agents to assess the effects of reform. Reforms may increase uncertainty and dampen the response of economic agents, particularly investors, to the reform program.

Closely related to the issue of comprehensiveness is that of the pace of reform. Again, Stuart, while noting there are no easy answers, believes that quicker is better than slower. This presumption is subject to qualifications like those guarding the argument for comprehensiveness. Indeed, as Stuart recognizes, there is a trade-off between the quantity and quality of fiscal adjustment, and quality often argues for gradualism. Similar quality and quantity trade-offs arise in other spheres of adjustment.

In the messy real world, programs are often designed with tight deadlines, invariably with imperfect information, and necessarily under the exercise of judgment. The questions are how to narrow the area of judgment and which policies should be given priority. To minimize the scope for judgment and translate general lessons into the context of a particular country—or discover the mutatis mutandis that attaches to any general lesson—is more challenging in the absence of good policy-relevant research on the economy at hand. Usually such research is scarce in developing countries. In these circumstances, the temptation to rely excessively on general lessons must be stoutly resisted.

In Pakistan, economic policymaking more often than not must be undertaken with a paucity of information and policy-relevant research. And Pakistan, in my view, is better than other developing countries on this count. This lack of analysis and inadequacy of data frequently reflect the fact that policymakers have not demanded them consistently or for a long enough period of time. Statistical systems, a capacity for policy analysis and research, cannot be created overnight.

Thus, policymakers can sometimes find themselves the victims of the past policymaking mistakes.

Outside agencies, like the IMF or the World Bank, can only partly substitute for the weakness of a domestic statistical system. While the IMF is often criticized for its policy presumptions and conditionalities, countries should perhaps work to overcome their own vacuum of analysis before criticizing too loudly. Too often, critics deride a package of adjustment policies without offering any clear alternative or recognizing, as Stuart points out, that in the face of macroeconomic imbalance, countries have to adjust. The choice frequently is not between an adjustment package and its absence but between orderly and disorderly adjustment. Critics of IMF programs in Pakistan often tend to focus on the economic and social costs of an adjustment program but not on the costs of not adjusting at all.

The IMF has been criticized for not paying enough attention to minimizing the costs of adjustment and for doing too little too late on the supply side and the poverty alleviation issue. Whatever the merits of such a charge, it should also be recognized that the problem often lies with governments. Authorities frequently tend to be more concerned with the costs imposed on the politically powerful than on the really poor.

Stuart neglects or is perhaps too polite to mention the difficulties that such country “circumstances” present to the design of appropriately country-specific programs. As a result, the pressure to lean on general lessons becomes harder to resist. Some of these lessons are more contentious and less general than Stuart seems to suggest in his useful and stimulating paper.

1

This paper is based in part on the work of International Monetary Fund staff involved in the ongoing review of the experience of countries with IMF-supported programs. Views expressed in this paper are the sole responsibility of the author and do not necessarily reflect those of the IMF.

2

For a discussion of the relationship between demand-oriented fiscal and structural measures, see Hernandez-Cata (1989).

3

See Guitian (1981).

4

An exception is found in Khan and Knight (1985). The authors estimate a simple econometric model for developing countries to evaluate the effects of stabilization and structural reform measures. They conclude that once the influence of relevant policies on the growth rate is recognized, there is no clear presumption that IMF-supported adjustment programs adversely affect growth, even in the short run.

5

Khan (1990).

6

Heller and others (1988).

7

See Aghevli and others (1990).

8

See Tanzi (1989).

1

The author is on leave from the World Bank and serving as an economic advisor to Pakistan’s Ministry of Finance. The views expressed in this paper are the author’s own and should not be attributed to any institution.

2

The tendency to exaggerate what programs can achieve may arise because of either a self-serving bias on the part of technocrats involved in the program or pressures to display “viable” programs to executive boards and management of international organizations or to the domestic political leadership.

Reference

  • Aghevli, Bijan B., and others, The Role of National Saving in the World Economy Occasional Paper No. 67 (Washington: International Monetary Fund, 1990).

    • Search Google Scholar
    • Export Citation
  • Guitian, Manuel, “Fund Conditionality: Evolution of Principles and Practices,” IMF Pamphlet Series No. 38 (Washington: International Monetary Fund, 1981).

    • Search Google Scholar
    • Export Citation
  • Heller, Peter S., and others, The Implications of Fund-Supported Adjustment Programs for Poverty, Occasional Paper No.58 (Washington International Monetary Fund, 1988).

    • Search Google Scholar
    • Export Citation
  • Hernandez-Cata, Ernesto, “Issues in the Design of Growth Exercises: The Role of Fiscal and Structural Policies,” Staff Studies for the World Economic Outlook (Washington: International Monetary Fund, 1989).

    • Search Google Scholar
    • Export Citation
  • Khan, Mohsin S., “The Macroeconomic Effects of Fund-Supported Adjustment Programs,” Staff Papers, International Monetary Fund, Vol. 37 (June 1990), pp. 195231.

    • Search Google Scholar
    • Export Citation
  • Khan, Mohsin S., and Malcolm D. Knight, Fund-Supported Adjustment Programs and Economic Growth, Growth, Occasional Paper No.41 (Washington: International Monetary Fund, 1985).

    • Search Google Scholar
    • Export Citation
  • Tanzi, Vito, “Fiscal Policy, Growth, and the Design of Stabilization Programs,” in Fiscal Policy, Stabilization, and Growth in Developing Countries, edited by Mario I. Blejer and Ke-young Chu (Washington: International Monetary Fund, 1989).

    • Search Google Scholar
    • Export Citation

List of Participants

  • Moderator

    • V.A. Jafarey

    • Karachi

  • Authors

    • Qazi M. Alimullah

    • Secretary, Ministry of Finance

    • Government of Pakistan

    • Ibrahim Elwan

    • Manager, Co-Financing/Financial Advisory Service

    • World Bank

    • Mohsin S. Khan

    • Senior Advisor, Research Department

    • International Monetary Fund

    • Brian C. Stuart

    • Senior Advisor, Western Hemisphere Department

    • International Monetary Fund

    • Venkataraman Sundararajan

    • Deputy Director, Monetary and Exchange Affairs Department

    • International Monetary Fund

    • Vito Tanzi

    • Director, Fiscal Affairs Department

    • International Monetary Fund

  • Discussants

    • I.A. Hanfi

    • Governor, State Bank of Pakistan

    • Karachi

    • Malcolm D. Knight

    • Assistant Director, Research Department

    • International Monetary Fund

    • Azizali F. Mohammed

    • Alternate Executive Director

    • International Monetary Fund

    • Akbar Noman

    • Director, Economic Policy Research Unit

    • Lahore

    • Balwanth Reddy

    • Administrative Staff College of India

    • Hyderabad

  • Participants

    • Ahmed Abushadi

    • Senior Information Officer, External Relations Department

    • International Monetary Fund

    • Mueen Afzal

    • Additional Secretary, Ministry of Finance

    • Government of Pakistan

    • Khorshed Alam

    • Principal Finance Secretary, Ministry of Finance

    • Government of Bangladesh

    • Syed Babar Ali

    • Chairman, Packages (Pvt.) Ltd.

    • Lahore

    • Sayyida Al-Jalali

    • Professor, Center for Applied Economic Studies

    • Peshawar University

    • Peshawar

    • M. Arif

    • Chairman, Planning and Development Department

    • Government of Punjab

    • Lahore

    • Viqar Rustam Bakhahi

    • Additional Secretary (Dev.), Ministry of Industries

    • Government of Pakistan

    • Tariq Binori

    • Environmental Policy Advisor

    • International Union for Conservation of Nature and Natural

    • Resources

    • Islamabad

    • S.B. Chaudhuri

    • Governor, Bangladesh Bank

    • Dhaka

    • Razaq Dawood

    • Managing Director, Descon Engineering Ltd.

    • Lahore

    • Julio De Quesada

    • General Manager (Corporate), Citibank N.A.

    • Karachi

    • Hasan Ersel

    • Vice Governor, Central Bank of Turkey

    • Ulus/Ankara

    • Mohamed Finaish

    • Executive Director

    • International Monetary Fund

    • Akmal Hussain

    • Managing Director, Syed Engineering Co.

    • Lahore

    • M. Ashraf Janjua

    • Economic Advisor, State Bank of Pakistan

    • Karachi

    • P.B. Jayasundera

    • Additional Director, Fiscal Policy Division, General Treasury

    • Government of Sri Lanka

    • A.R. Kamal

    • Pakistan Institute of Development Economics

    • Quaid-e-Azam University

    • Islamabad

    • Shahid Kardar

    • Systems Ltd.

    • Lahore

    • A.H. Khan

    • Secretary, Planning and Development, Pakistan Secretariat

    • Government of Pakistan

    • Alia Khan

    • Assistant Professor, Economics Department

    • Quaid-e-Azam University

    • Islamabad

    • Kursheed A. Khan

    • President, All Pakistan Federation of Trade Unions

    • Lahore

    • Shamim Ahmad Khan

    • Corporate Law Authority

    • Government of Pakistan

    • W.D. Lakshman

    • Professor, Department of Economics, University of Colombo

    • Colombo

    • Akhtar Mahmood

    • Programme Specialist

    • Agriculture and Rural Development Department

    • Government of Pakistan

    • Iqbal Mueen

    • Additional Secretary, Ministry of Commerce

    • Government of Pakistan

    • Kassim Parekh

    • Advisor, Habib Bank, A.G. Zurich

    • Karachi

    • Kirit Parikh

    • Director, Indira Gandhi Institute for Economic Research

    • Bombay

    • Hafeez Pasha

    • Director, Institute of Applied Economics, Karachi University

    • Karachi

    • Shakour A. Shaalan

    • Director, Middle Eastern Department

    • International Monetary Fund

    • Ibrahim Shaibani

    • Vice Governor, Central Bank of the Islamic Republic of Iran

    • Tehran

    • Yousuf Shirazi

    • Chairman, Atlas Group of Companies

    • Karachi

    • Fasih Uddin

    • Economic Advisor, Finance Division

    • Government of Pakistan

    • Arshad Zaman

    • Chief Economist/Special Secretary

    • Planning Division

    • Government of Pakistan

    • M. Ziauddin

    • Bureau Chief, Daily Dawn

    • Islamabad

    • Khaliq Zuberi

    • Editor (Business), Daily News

    • Karachi